Aggregating intrinsic information to enhance BCI performance through federated learning

Rui Liu, Yuanyuan Chen, Anran Li, Yi Ding, Han Yu, Cuntai Guan

Abstract: Highlights•This work provided new insights on federated learning for the BCI.•Enabled knowledge sharing across heterogeneous EEG datasets.•Proposed a new paradigm to build a deep learning model for BCI.•Designed a personalized FL EEG decoding framework, evaluated on 9 MI datasets.•Improved classification performance up to 8.4% by the proposed FL framework.